DIFFERENCES OF ATTACHMENT STYLES TO FATHERS BETWEEN EARLY ADOLESCENT MALE AND FEMALE

Many opinions state that girls have a tendency to be closer to their fathers than boys. This is because fathers usually reveal soft attitude and a kind way of talking toward his daughter. In addition, fathers also succumb to the demands of his daughter. Conversely, with his son, fathers are usually much more 'rough and tough'. So, girls feels more secure to their fathers than boys. The objective of this study was to determine the differences of attachment style to fathers between early adolescent male and female in seventh grade classes of Junior High School 19 Banda Aceh. The number of samples taken were 60 students. A rating scale was used for collecting data, namely Adult Attachment Scale (AAS). The result obtained from the T-Tests showed there was no significant differences of attachment style to fathers between early adolescents male and female in seventh grade classes of Junior High School 19 Banda Aceh. However, early adolescents prefer to have close attachment style with their fathers.

developmental scientists have shown more interest in autonomy rather than attachment during adolescence.
Attachment was originally developed in infancy, where mother became the primary attachment figure in the process of establishing child's emotion regulation (Mikulincer & Shaver, 2007).According to Santrock (2011), attachment to the mother in adolescence will help the establishment of competence and social welfare.This is reflected in the characteristics of self-esteem, emotional adjustment, and physical health.
But, with a change of paradigm, father has an important role in the formation of attachment (Doyle et al., 2009).
Changing of paradigm emphasizes that in addition to be a main breadwinner in the family, the father also has an important role in the formation of behavior in adolescents.In a patriarchal culture, the father has a dominant role over the mother.The father is head of the family, who must be responsible to personal, wife, children, family, and society (Bretherton, 2010).
Aceh is one of the states in Indonesia that has a patriarchal culture, where the father becomes the central decision maker in family, and has no interest in household activities.Social construct of the local community put the wife (female) on the nature of domestic work (household work), while the husband (male) do more work outside the home.Meaning, based on social construct, the husband who wants to do housework is a bonus for his wife, because he wants to do work that does not belong to his responsibility ( Srimulyani & Inayatillah, 2009).
However, a new phenomenon was found after the 2004 Tsunami disaster.A lot of foreign and domestic NGOs came to Aceh to carry out rehabilitation and reconstruction process in the framework of the economic recovery.In any activities, many NGOs pay attention to the aspect of gender equality, especially in activities to increase knowledge and skills through training in small business management and provision of working capital (Lisna, Agussabti, & Safrida, 2011).So, today wives have greater contribution outside the house which increases economic condition of the family.In some cases, especially in Aceh, it was found that when contribution of women to domestic economy strengthened, the husband was voluntarily willing to do housework.Before the Tsunami, that was not part of the husbands work (Lisna et al., 2011).It could be argued that although Aceh culture emphasized the importance of the mother role in educating children, you may find that now the father has taken the greater role of the mothers duties.Because of this, attachment to the father has a great contribution to the development of adolescents personality.
A study conducted by Akhtar (2012) found that the father trained male adolescents for performing out of home tasks, so they spent more time outside, in comparrison to female adolescents.The father's strictness forces them to avoid and not to indulge themselves in the wrong activities.In addition, the father favors the female child and keeps strict check on the male child.

Literature Review
A study showed that the father attachment was more significantly related to social competence in various social situations (Ma & Huebner, 2008).
Additionally, Santrock (1996) also said that men and women who had good social relationships (marital quality and ties outside the family, such as friendship), will experience more warmth from the father during childhood (Santrock, 1996).Besides that, a recent study conducted by Fosco (2012) indicated that parental monitoring and father-youth connectedness were associated with reductions in problem behavior over time.
Attachment given by the father to the son since infancy, will give a feeling of comfort and love until he becomes an adult (Sullivan, 1999).The Father is the greatest asset in educating the emotions of early adolescent males, by giving full attention to the son.The father also provides positive benefits for the adolescent son, such as in the development of self-control, the ability to delay the desire, and helps adolescents in social adjustment.
Daughter-to-father attachment has been shown to be significantly related to a daughter's ability to regulate emotions (Pearce, 2009).Kenny and Gallagher (2002) also said that paternal attachment is associated positively and significantly with traditional feminine traits of expressivity.Expressivity may be linked with emotional regulation in females who are expressive.Females may be able to better articulate their emotions as compared to those females who may be more withdrawn.In addition, paternal involvement may be a protective factor against psychological male adjustment in adolescents with families that are not intact and against psychological distress, especially for women (Flouri & Buchanan, 2003).

Secure Attachment
Secure attachment is a attachment style characterized by children who feel comfortable with their parents, are not completely dependent on their parents, have fun relationships with their parents, and have a sense of confidence.When entering adulthood, those who obtain secure attachment tend to have a sense of trust, long term relationships, have high self-esteem, enjoying intimate relations, seek social support, and have the ability to share feelings with others (Kobak, Rosenthal, Zajac, & Madsen, 2007).In addition, securely attached adolescents exhibited more positive affect and orientation in problem solving-tasks (Easterbrooks & Goldberg, 1984).

Anxious Attachment
Anxious attachment is an attachment style characterized by parents who are not sensitive to their children, less responsive to the needs of their children, and act unfair.When entering adulthood, those who get anxious attachment often feel reluctant to close relationships with others and worry that their partner does not return their feelings.This often causes broken, cold, and distant relationships.These people feel very sad when the relationships end (Mikulincer & Shaver, 2007).

Avoidant Attachment
Avoidant attachment is an attachment style characterized by parents who do not pay attention and have suffered rejection.The children also have cold relations with parents, can feel uncomfortable when being with their parents, do not want established close relationships with their parents, and have difficulty trusting their parents.These individuals do not invest much emotion in relationships and experience little pain when the relationships end (Cassidy, 1994).

Objective
The objective of this study was to determine the differences of attachment styles to fathers between early adolescent males and females.

Research Design
The approach used was a quantitative method.This is because a quantitative approach has many advantages for this paticular study, such as providing systematic process to obtain quantifiable information and presenting in numerical form.In addition, quantitative approach also uses a statistical method that emphasizes the analysis of the numerical data (numbers).

Procedure
Samples were taken using random sampling with proportionate stratified random sampling technique.The number of samples taken was stratified because the population was also stratified, based on the classes that consist of VII-1, VII-2, VII-3, and VII-4.

Participants
Population in this study was seventh grade students of Junior High School 19 Banda Aceh that meet the established criteria, specifically at least 12 years old and still have a father.So, the number of samples taken were 60 students with 30 males and 30 females.

Data Collection Method
The method used in this study was a survey method which took samples from a certain population and used questionnaires as the main data collection tool (Singarimbun & Effendi, 1989).The primary data obtained from the respondents were through a questionnaire.In the filling process, the author handed the questionnaire directly to the respondent and returned immediately after being filled.This method selected to get more certainty in obtaining the data, saving time and cost as well.

Operational Definitions
Attachment style to fathers are a style of emotional bond between seventh grade students of Junior High School 19 Banda Aceh and their fathers.Emotional bonds encourage giving security, protection, and safety in order to develop positive social relationships.And, early adolescent is defined as an adolescent who is at least 12 years old and a maximum of 15 years old.

Demographic Information
A brief demographic questionnaire was used to obtain background characteristics of the participants.Items in this questionnaire would request information on gender, age, and number of siblings.

Attachment Style to Father
A rating scale was used in collecting data, namely Adult Attachment

Validity Test
Statistical technique used to find the correlation coefficient was the Pearson Product Moment Correlation technique.In this case, the calculation of the items validity would be analyzed by the SPSS program.
Below is the formula : rxy Description : rxy is correlation coefficient, X is total score of each item of all tryout respondents, Y is total score of entire item of all try-out respondents, and n is total of try-out respondents (Arikunto, 2000).
The calculation results would be compared with the table of r values in correlation with significant levels of 5%.If the calculation of Pearson Product Moment Correlation is above the r values at significant level of 5%, then the items has good validity (Sugiyono, 1999).

Reliability Test
Reliability is the extent to which the results of the measurement can be trusted (Azwar, 1999).Reliability estimation technique used was the Cronbach's Alpha coefficient technique, with the following formula.
Description : r11 is scale reliability, k is total of items, is total of items variance, and is sum of variance total.
Ghozali ( 2001) said that a construct or variable could be said reliable if value of Cronbach's Alpha > .60.

Normality Test
Normality test is used to consider the inductive statistics that would be performed, whether parametric statistics or non-parametric statistics.If hypothesis testing is using parametric statistics, the variable data needs to be assessed by the normality test (Hadi, 2000).
The Normality test used was one sample, the Kolmogorov-Smirnv test.This test was used to prove whether the data has followed the normal distribution or not.This test used a two-tailed significant level, if the results were greater than a significant level, the data was normally distributed.Conversely, if the results were less than the significant level, the data was not normally distributed.

Homogeneity of Variance Test
Homogeneity of the variance test is useful to know because the variable in one group may not be significantly different than another (homogeneous).This test was only performed on normal distributed data.

Homogeneity of variant test used was Levene's Test of Equality of
Variance.The values results were the probability of a two-tailed, which then compared with the significant level.If the value is greater than the probability of significant level, then the variance between groups was homogeneous or groups of variables had the same variance.Otherwise if the value was less than the probability of significant level, then the variables between groups were heterogeneous.

Independent Sample T-Tests
This test was used to assess the differences of mean or average values of two populations.This is based on the assumption that the population variance must be identical or similar to the population where the samples were taken and should be normally distributed.T-tests are based on positive or negative signs from the differences between pair observation, not based on the amount of the differences.

Validity Test
In this study, the author used item analysis by examining the correlation between the score of the item and the total score.The correlation coefficient was obtained by using Pearson Product Moment Correlation.The results of validity test were presented in Table 1.The analysis showed that all items positively correlated to the total score.The conclusion was all items in the instrument used to measure the Attachment Style to the Father, were valid because they are larger than an r-table of .26.

Reliability Test
Testing was done using Cronbach's Alpha and the result obtained was of .95.This value showed that the items in the instrument were reliable, because r11 > .60.Because the data was valid and reliable, they can be used for hypothesis testing.

Normality Test
Normality test was done to determine whether the population followed normal distribution or not.Normality testing technique used in this study was Kolmogorov Smirnov Test.The results could be seen in table 2 below.The results of the normality test concluded that the data was normally distributed, because the significant value > .05.Because the samples were normally distributed, the statistical test used was parametric statistical Independent Sample T-Tests.

Homogeneity of Variance Test
Results obtained from this test was the probability for two-tailed, which then compared with significant level in amount of .05.Calculation results can be seen in Table 3.

.02 Heterogeneous
From Table 3, it was known that the probability value for Attachment Style to the Father variable, shows heterogeneous data (p < .05).Because the average value of the probability was less than the significant level of .05, the variance between groups was heterogeneous and the assumption used was different variance assumption (Equal variances not assumed).

Independent Sample T-Tests
The hypothesis was tested by using Independent Sample T-Tests.
The results of the Independent Sample T-Tests, with different variance assumption (Equal variances not assumed), could be seen in Table 4. Results indicated a non-significant trending in attachment style to the father for early adolescent females (M = 62.63,SD = 12.99), and over early adolescent males (M = 59.20,SD = 9.26), t(52.41)= -1,18, p = .24.This result supports the research findings by Akhtar (2012).Akhtar conducted a study toward the adolescents that have an age ranged from 15 to 18 years old, and consisting of 98 males and 81 females.The findings showed no significant differences were found in attachment styles of male and female respondents.
However, close attachment on early adolescents female had a mean value of 21.50 (SD = 2.96) and were respectively higher than depend (M = 20.63,SD = 2.99), t(50.84)= -1.37,p = .18and anxiety (M = 20.50,SD = 3.88), t(55.88)= -.95,p = .34.In addition, close attachment on early adolescents male had a mean value of 20.47 (SD = 4.3) and were also respectively higher than depend (M = 19.30,SD = 4.43), t(50.84)= -1.37,p = .18and anxiety (M = 19.43,SD = 4.73), t(55.88)= -.95,p = .34.This indicated the preference of female and male respondents for close attachment style, because their analytical thinking is still immature.Most of them do not have enough knowledge to make appropriate decisions in all areas of life.When early adolescents demand their autonomy by trying to solve their own problems, the wise father will guide them to make reasonable decisions in areas where their knowledge is still limited (Santrock, 2011).This guidance indicates the form of close attachment between fathers and adolescents.

D. Discussions
The result from Independent Sample T-Test for AAS subscales showed that there was no significant difference in attachment styles to the father between early adolescent males and females at seventh grade classes of Junior High School 19 Percontohan Banda Aceh.This result supported the research finding by Akhtar (2012) who conducted a study with adolescents with an age range from 15 to 18 years old.The study consisted of 98 male and 81 female.The findings showed no significant differences in attachment styles of male and female respondents.Besides that, many researchers also have found no sex differences in attachment styles, when using interview-based assessments of attachment (van IJzendoorn & Bakermans-Kranenburg, 1996).
Sex differences were not a major factor to determine the differences in developing attachment styles to the father.Akhtar (2012) said that there are many factors which contribute in developing attachment styles, but the most important is the parent-child relationship.Parent-child relations might predict the outcome of the child's future in the interpersonal relationships.Other factors that might contribute to attachment style to the father are individual differences and parenting styles.This is also consistent with the expression of Ainsworth (1968), although every child gets intensive attachment since the first mid-year, a child does not necessarily show the same attachment with the other children.
Parenting style applied by parents will also largely determine the quality of attachment.Parents who are sensitive and always ready to respond to the needs of their children will provide a secure attachment for the child.

E. Conclusion and Implication
The conclusion from this study was that there are no significant differences between early adolescents males and females in regards to their attachment to their father.It can be said that boys feel secure and attached to their father as well as daughters.Therefore, fathers should be able to maintain a secure relationship with adolescents by understanding their needs and characters.So, the children can sustain a close relationship with their father.

F. Recommendations
Another area which is needed to be investigated in future research is early adolescents' relationships with multiple attachment figures.
Examining relationships with each attachment figure (e.g., mother, samesex friend, and romantic friend) may enrich our understanding of the multiple relationships integral to development.This would also be appropriate for studying adolescents from diverse family structures and different ethnic and cultural backgrounds, who may have close relationships with a number of people besides biological parents and close friends.Other variables related to attachment styles to father, such as social skills, personality characteristics, birth order, and child-parent relationship should also be explored.

G. Limitations
The first limitations of this study was that the results could not be generalized to a broader scope, because the study only conducted on students at Junior High School 19 Banda Aceh.The following was, the method used to collect data.If qualitative method was used with performing interviews and observation, perhaps the data would be more detail and comprehensive.
Scale (AAS) by Collins and Read.Before distributing, this scale was changed into Indonesian Language, because most of participants speak this language in daily life.This scale consist of three dimensions of attachment styles (close attachment, depend attachment, and anxiety attachment), in which each dimensions has 6 items.Respondents are asked to choose one of four alternative options, from Strongly Inappropriate (SI), Inappropriate (I), Appropriate (A), and Strongly Appropriate (SA).For favorable item (+), Strongly Appropriate (SA) got a score of 4, Appropriate (A) got a score of 3, Inappropriate (I) got a score of 2, and Strongly Inappropriate (SI) got a score of 1. Otherwise, for unfavorable item (-), Strongly Appropriate (SA) got a score of 1, Appropriate (A) got a score of 2, Inappropriate (I) got a score of 3, and Strongly Inappropriate (SI) got a score of 4. Score for each item is given by suitability between answer and attachment style to the father which showed in the related item.

Table 1 .
Validity Test of Adult Attachment Scale

Table 3 .
Homogeneity of Variant Test

Table 4 .
Independent Sample T-Tests